view tests/test-pull-update.t @ 38732:be4984261611

merge: mark file gets as not thread safe (issue5933)

In default installs, this has the effect of disabling the thread-based
worker on Windows when manifesting files in the working directory. My
measurements have shown that with revlog-based repositories, Mercurial
spends a lot of CPU time in revlog code resolving file data. This ends
up incurring a lot of context switching across threads and slows down
`hg update` operations when going from an empty working directory to
the tip of the repo.

On mozilla-unified (246,351 files) on an i7-6700K (4+4 CPUs):

before: 487s wall
after:  360s wall (equivalent to worker.enabled=false)
cpus=2: 379s wall

Even with only 2 threads, the thread pool is still slower.

The introduction of the thread-based worker (02b36e860e0b) states that
it resulted in a "~50%" speedup for `hg sparse --enable-profile` and
`hg sparse --disable-profile`. This disagrees with my measurement
above. I theorize a few reasons for this:

1) Removal of files from the working directory is I/O - not CPU - bound
   and should benefit from a thread pool (unless I/O is insanely fast
   and the GIL release is near instantaneous). So tests like `hg sparse
   --enable-profile` may exercise deletion throughput and aren't good
   benchmarks for worker tasks that are CPU heavy.
2) The patch was authored by someone at Facebook. The results were
   likely measured against a repository using remotefilelog. And I
   believe that revision retrieval during working directory updates with
   remotefilelog will often use a remote store, thus being I/O and not
   CPU bound. This probably resulted in an overstated performance gain.

Since there appears to be a need to enable the thread-based worker with
some stores, I've made the flagging of file gets as thread safe
configurable. I've made it experimental because I don't want to formalize
a boolean flag for this option and because this attribute is best
captured against the store implementation. But we don't have a proper
store API for this yet. I'd rather cross this bridge later.

It is possible there are revlog-based repositories that do benefit from
a thread-based worker. I didn't do very comprehensive testing. If there
are, we may want to devise a more proper algorithm for whether to use
the thread-based worker, including possibly config options to limit the
number of threads to use. But until I see evidence that justifies
complexity, simplicity wins.

Differential Revision: https://phab.mercurial-scm.org/D3963

  $ hg init t
  $ cd t
  $ echo 1 > foo
  $ hg ci -Am m
  adding foo

  $ cd ..
  $ hg clone t tt
  updating to branch default
  1 files updated, 0 files merged, 0 files removed, 0 files unresolved
  $ cd tt
  $ echo 1.1 > foo
  $ hg ci -Am m

  $ cd ../t
  $ echo 1.2 > foo
  $ hg ci -Am m

Should respect config to disable dirty update
  $ hg co -qC 0
  $ echo 2 > foo
  $ hg --config commands.update.check=abort pull -u ../tt
  pulling from ../tt
  searching for changes
  adding changesets
  adding manifests
  adding file changes
  added 1 changesets with 1 changes to 1 files (+1 heads)
  new changesets 107cefe13e42
  1 local changesets published
  abort: uncommitted changes
  [255]
  $ hg --config extensions.strip= strip --no-backup tip
  $ hg co -qC tip

Should not update to the other topological branch:

  $ hg pull -u ../tt
  pulling from ../tt
  searching for changes
  adding changesets
  adding manifests
  adding file changes
  added 1 changesets with 1 changes to 1 files (+1 heads)
  new changesets 107cefe13e42
  0 files updated, 0 files merged, 0 files removed, 0 files unresolved
  updated to "800c91d5bfc1: m"
  1 other heads for branch "default"

  $ cd ../tt

Should not update to the other branch:

  $ hg pull -u ../t
  pulling from ../t
  searching for changes
  adding changesets
  adding manifests
  adding file changes
  added 1 changesets with 1 changes to 1 files (+1 heads)
  new changesets 800c91d5bfc1
  1 local changesets published
  0 files updated, 0 files merged, 0 files removed, 0 files unresolved
  updated to "107cefe13e42: m"
  1 other heads for branch "default"

  $ HGMERGE=true hg merge
  merging foo
  0 files updated, 1 files merged, 0 files removed, 0 files unresolved
  (branch merge, don't forget to commit)
  $ hg ci -mm

  $ cd ../t

Should work:

  $ hg pull -u ../tt
  pulling from ../tt
  searching for changes
  adding changesets
  adding manifests
  adding file changes
  added 1 changesets with 1 changes to 1 files (-1 heads)
  new changesets 483b76ad4309
  1 local changesets published
  1 files updated, 0 files merged, 0 files removed, 0 files unresolved

Similarity between "hg update" and "hg pull -u" in handling bookmark
====================================================================

Test that updating activates the bookmark, which matches with the
explicit destination of the update.

  $ echo 4 >> foo
  $ hg commit -m "#4"
  $ hg bookmark active-after-pull
  $ cd ../tt

(1) activating by --rev BOOKMARK

  $ hg bookmark -f active-before-pull
  $ hg bookmarks
   * active-before-pull        3:483b76ad4309

  $ hg pull -u -r active-after-pull
  pulling from $TESTTMP/t
  searching for changes
  adding changesets
  adding manifests
  adding file changes
  added 1 changesets with 1 changes to 1 files
  adding remote bookmark active-after-pull
  new changesets f815b3da6163
  1 local changesets published
  1 files updated, 0 files merged, 0 files removed, 0 files unresolved
  (activating bookmark active-after-pull)

  $ hg parents -q
  4:f815b3da6163
  $ hg bookmarks
   * active-after-pull         4:f815b3da6163
     active-before-pull        3:483b76ad4309

(discard pulled changes)

  $ hg update -q 483b76ad4309
  $ hg rollback -q

(2) activating by URL#BOOKMARK

  $ hg bookmark -f active-before-pull
  $ hg bookmarks
   * active-before-pull        3:483b76ad4309

  $ hg pull -u $TESTTMP/t#active-after-pull
  pulling from $TESTTMP/t
  searching for changes
  adding changesets
  adding manifests
  adding file changes
  added 1 changesets with 1 changes to 1 files
  adding remote bookmark active-after-pull
  new changesets f815b3da6163
  1 local changesets published
  1 files updated, 0 files merged, 0 files removed, 0 files unresolved
  (activating bookmark active-after-pull)

  $ hg parents -q
  4:f815b3da6163
  $ hg bookmarks
   * active-after-pull         4:f815b3da6163
     active-before-pull        3:483b76ad4309

(discard pulled changes)

  $ hg update -q 483b76ad4309
  $ hg rollback -q

Test that updating deactivates current active bookmark, if the
destination of the update is explicitly specified, and it doesn't
match with the name of any existing bookmarks.

  $ cd ../t
  $ hg bookmark -d active-after-pull
  $ hg branch bar -q
  $ hg commit -m "#5 (bar #1)"
  $ cd ../tt

(1) deactivating by --rev REV

  $ hg bookmark -f active-before-pull
  $ hg bookmarks
   * active-before-pull        3:483b76ad4309

  $ hg pull -u -r b5e4babfaaa7
  pulling from $TESTTMP/t
  searching for changes
  adding changesets
  adding manifests
  adding file changes
  added 2 changesets with 1 changes to 1 files
  new changesets f815b3da6163:b5e4babfaaa7
  1 local changesets published
  1 files updated, 0 files merged, 0 files removed, 0 files unresolved
  (leaving bookmark active-before-pull)

  $ hg parents -q
  5:b5e4babfaaa7
  $ hg bookmarks
     active-before-pull        3:483b76ad4309

(discard pulled changes)

  $ hg update -q 483b76ad4309
  $ hg rollback -q

(2) deactivating by --branch BRANCH

  $ hg bookmark -f active-before-pull
  $ hg bookmarks
   * active-before-pull        3:483b76ad4309

  $ hg pull -u -b bar
  pulling from $TESTTMP/t
  searching for changes
  adding changesets
  adding manifests
  adding file changes
  added 2 changesets with 1 changes to 1 files
  new changesets f815b3da6163:b5e4babfaaa7
  1 local changesets published
  1 files updated, 0 files merged, 0 files removed, 0 files unresolved
  (leaving bookmark active-before-pull)

  $ hg parents -q
  5:b5e4babfaaa7
  $ hg bookmarks
     active-before-pull        3:483b76ad4309

(discard pulled changes)

  $ hg update -q 483b76ad4309
  $ hg rollback -q

(3) deactivating by URL#ANOTHER-BRANCH

  $ hg bookmark -f active-before-pull
  $ hg bookmarks
   * active-before-pull        3:483b76ad4309

  $ hg pull -u $TESTTMP/t#bar
  pulling from $TESTTMP/t
  searching for changes
  adding changesets
  adding manifests
  adding file changes
  added 2 changesets with 1 changes to 1 files
  new changesets f815b3da6163:b5e4babfaaa7
  1 local changesets published
  1 files updated, 0 files merged, 0 files removed, 0 files unresolved
  (leaving bookmark active-before-pull)

  $ hg parents -q
  5:b5e4babfaaa7
  $ hg bookmarks
     active-before-pull        3:483b76ad4309

  $ cd ..